网络模糊理论在有功静态安全域中的应用

本文将网络模糊理论用于电力系统有功静态安域计算，在模型中引入了线路短时过载的失效概率和模糊线路功率限制约束，分析和讨论了不同隶属度下有功静态安全域的大小。

基于模糊理论与神经网络的建筑电气设备故障诊断研究

本文简要介绍了建筑电气常见故障类型和危害,阐述了模糊理论与神经网络的基本原理,根据建筑电气设备的故障诊断采用的技术,提出来将模糊理论与神经网络这2个基本理论结合在一起的故障诊断模型,探讨将该模型应用于建筑电气设备中的故障诊断研究,这一理论的提出也会为建筑电气设备故障诊断的发展打开一条全新的思路。

基于Fuzzy-ART神经网络的红外弱小目标检测

针对现有背景抑制算法未能有效抑制背景而导致目标检测率低的问题,提出了一种基于模糊自适应共振理论(fuzzy adaptive resonance theory,Fuzzy-ART)神经网络的弱小目标检测算法。首先,采用Fuzzy-ART神经网络结合Robinson警戒环技术,建立自适应局部空间背景模型,并以此分析像素点的背景模糊隶属度来抑制背景杂波;然后依据目标与残留背景杂波的空间特征采用模板均差法来突显目标,并提出基于行列模糊聚类的自适应分割算法来提取候选目标;最后结合目标的运动连续性进行多帧轨迹关联从而检测出真实目标。理论分析与实验结果表明,该算法能随背景的局部情况来自适应调节空间背景模型,从而自适应抑制背景杂波、突显目标,能有效提高信噪比,检测出弱小目标。

FART神经网络的改进及其在晶圆在线监测中的应用

对模糊自适应共振理论(FART)神经网络进行改进,使用改进的FART神经网络对半导体生产线晶圆合格率进行在线检测,对晶圆合格率特征向量进行聚类分析,将合格率损失中拥有相类似特征的晶圆分为一类,一旦检测到生产线发生异常,便可找出故障设备并及时维护,从而使生产线处于高生产率状态。

FUZZY ARTMAP在三字词声调识别中的应用

三字词音节声调模式具有连续语音中音节声调模式的特征,声调的提取和识别远较孤立字困难。采用小波变换方法提取语音基音,用Fuzzy ARTMAP神经网络进行声调识别,获得了比BP网络更好的实验结果。分析了仿真参数对识别结果的影响,讨论了Fuzzy ARTMAP神经网络中的过拟合问题,给出了一种基于Fuzzy AR-TMAP神经网络的三字词声调识别方法。

智能MAS在分布式远程教学系统中的应用研究

智能多代理系统(Multi-Agent System,MAS)和模糊理论都是人工智能领域的研究热点,教学系统中的子系统 Agent可以作为网络教学管理员、教师、甚至作为教学对象陪伴学生学习。在基于 MAS 的分布式网络教学系统中引入模糊理论,可以实现优势互补,在分布式结构的基础上实现智能教学、学习评估,对网络教学技术的发展具有重要的现实意义。

Trust Management Mechanism for Internet of Things

Trust management has been proven to be a useful technology for providing security service and as a consequence has been used in many applications such as P2P, Grid, ad hoc network and so on. However, few researches about trust mechanism for Internet of Things （IoT） could be found in the literature, though we argue that considerable necessity is held for applying trust mechanism to IoT. In this paper, we establish a formal trust management control mechanism based on architecture modeling of IoT. We decompose the IoT into three layers, which are sensor layer, core layer and application layer, from aspects of network composition of loT. Each layer is controlled by trust management for special purpose： self-organized, affective routing and multi-service respectively. And the final decision-making is performed by service requester according to the collected trust information as well as requester＇ policy. Finally, we use a formal semantics-based and fuzzy set theory to realize all above trust mechanism, the result of which provides a general framework for the development of trust models of IoT.

A DATA MINING METHOD BASED ON CONSTRUCTIVE NEURAL NETWORKS

In this letter,Constructive Neural Networks (CNN) is used in large-scale data mining. By introducing the principle and characteristics of CNN and pointing out its deficiencies,fuzzy theory is adopted to improve the covering algorithms. The threshold of covering algorithms is redefined. "Extended area" for test samples is built. The inference of the outlier is eliminated. Furthermore,"Sphere Neighborhood (SN)" are constructed. The membership functions of test samples are given and all of the test samples are determined accordingly. The method is used to mine large wireless monitor data (about 3×107 data points),and knowledge is found effectively.

Simultaneous Perturbation Stochastic Approximation Algorithm Combined with Neural Network and Fuzzy Simulation

In order to solve three kinds of fuzzy programm model, fuzzy chance-constrained programming mode ng models, i.e. fuzzy expected value and fuzzy dependent-chance programming model, a simultaneous perturbation stochastic approximation algorithm is proposed by integrating neural network with fuzzy simulation. At first, fuzzy simulation is used to generate a set of input-output data. Then a neural network is trained according to the set. Finally, the trained neural network is embedded in simultaneous perturbation stochastic approximation algorithm. Simultaneous perturbation stochastic approximation algorithm is used to search the optimal solution. Two numerical examples are presented to illustrate the effectiveness of the proposed algorithm.

Comprehensively Context-Aware Approach to Guaranteeing Multimedia Conferencing Services

Service-Oriented Communication(SOC)is a key research issue to enable media communications using the Service-Oriented Architecture(SOA).Motivated by the necessity to guarantee the service quality of our webbased multimedia conferencing system,we present a Comprehensively Context-Aware(CoCA)approach in this paper.One major problem in the existing end-to-end Quality of Service(QoS)management solutions is that they analyse and exploit the relationships between the QoS metrics and corresponding contexts in an isolated manner.In this paper,we propose a novel approach to leveraging such relationships in a comprehensive manner based on Bayesian networks and the fuzzy set theory.This approach includes three phases:1)information feedback and training,2)QoS-to-context mapping,and3)optimal context adaption.We implement the proposed CoCA in the real multimedia conferencing system and compare its performance with the existing bandwidth aware and playback buffer aware schemes.Experimental results show that the proposed CoCA outperforms the competing approaches in improving the average video Peak Signal-to-Noise Ratio(PSNR).It also exhibits good performance in preventing the playback buffer starvation.

Memory Chain in a Chaotic Autoassociative Neural Network

Memory chain is observed in a chaotic autoassociative neural network. The network recalls first stored pattern from a fragment of a memory, stays at this pattern for a while, transits to the second stored pattern that overlaps with the first recalled pattern.Then it stays at the second recalled pattern for a while, transits to the third stored pattern that overlaps with the second recalled pattern, and so on. Thus a memory chain is generated. The memory chain ends with the pattern that overlaps no other stored patten. This phenomenon is similar to the way of recalling process of human beings in some respects.

Application of Fuzzy Automata Theory and Knowledge Based Neural Networks for Development of Basic Learning Model

The paper focuses on amalgamation of automata theory and fuzzy language. It uses adaptive knowledge based abstract framework which uses dynamic neural network framework along with fuzzy automata as Models of Learning, combining the two methodologies the authors develop a new framework termed as Fuzzy Automata based Neural Network （FANN）. It highlights conversion of knowledge rule to fuzzy automata thereby generating a framework FANN. FANN consists of composite fuzzy automation divided into ＂Performance Evaluator＂ and ＂Feature Extraction＂ which takes the help of previously stored samples of similar situations. The authors have extended FANN for Urban Traffic Modeling.

Theoretical Research on Novel Data Mining Algorithm based on Fuzzy Clustering Theory and Deep Neural Network

With the progress of computer technology, data mining has become a hot research area in the computer science community. In this paper, we undertake theoretical research on the novel data mining algorithm based on fuzzy clustering theory and deep neural network. The focus of data mining in seeking the visualization methods in the process of data mining, knowledge discovery process can be users to understand, to facilitate human-computer interaction in knowledge discovery process. Inspired by the brain structure layers, neural network researchers have been trying to multilayer neural network research. The experiment result shows that out algorithm is effective and robust.

Supply Chain Production-distribution Cost Optimization under Grey Fuzzy Uncertainty

Most supply chain programming problems are restricted to the deterministic situations or stochastic environmcnts. Considering twofold uncertainty combining grey and fuzzy factors, this paper proposes a hybrid uncertain programming model to optimize the supply chain production-distribution cost. The programming parameters of the material suppliers, manufacturer, distribution centers, and the customers are integrated into the presented model. On the basis of the chance measure and the credibility of grey fuzzy variable, the grey fuzzy simulation methodology was proposed to generate input-output data for the uncertain functions. The designed neural network can expedite the simulation process after trained from the generated input-output data. The improved Particle Swarm Optimization （PSO） algorithm based on the Differential Evolution （DE） algorithm can optimize the uncertain programming problems. A numerical example was presented to highlight the significance of the uncertain model and the feasibility of the solution strategy.

植基部落格之资料探勘模式以探索顾客需求

由于部落格(Blogs)的普及,导致愈来愈多的企业希望能从大量的使用者文章中撷取出有用的信息,从中了解消费者需求及市场导向,以帮助企业改善商品及服务质量,抑或评估企业本身或同业竞争者的优劣势。本研究针对部落格中的产品使用心得,提出一套FAIR模块,希望藉由该模块得以达到在短时间内有效地分析产品评价,以利于企业或消费者在掌握商品重点特色及整体评价时,能避免阅读大量文章的时间耗费并无从理出头绪的情形。FAIR模块为模糊自适应共振理论(Fuzzy ART)结合隐含语意索引(LSI)的特性,将文章集予以分群并从中撷取出代表性关键词,以达到信息检索的目的,最终再通过关联法则(AR)提升关键词的解释性。通过FAIR模块所撷取出来的消费者心声,我们更进一步地应用于质量机能展开,将顾客需求转化为技术需求,以分别了解产品本身或同业之间的竞争力,使企业充分掌握顾客需求,并提升产品设计之适用性。最后,我们以美容保养品之部落格文章作为实验对象,以说明并验证所提出的FAIR模块之效力。